Transform and quantization architecture for video coding and decoding

Abstract

A method of encoding a video stream in a video encoder is provided that includes computing an offset into a transform matrix based on a transform block size, wherein a size of the transform matrix is larger than the transform block size, and wherein the transform matrix is one selected from a group consisting of a DCT transform matrix and an IDCT transform matrix, and transforming a residual block to generate a DCT coefficient block, wherein the offset is used to select elements of rows and columns of a DCT submatrix of the transform block size from the transform matrix.

What is claimed is: 
     
       1. A method comprising:
 computing an offset into a transform matrix based on a transform block size, wherein a size of the transform matrix is larger than the transform block size; 
 selecting a submatrix of the transform block size from the transform matrix using the offset; 
 transforming an array of residual samples based on the submatrix to generate an array of transform coefficient values; and 
 selecting elements of rows and columns of the submatrix of the transform block size from the transform matrix using the offset. 
 
     
     
       2. The method of  claim 1 , further comprising computing an additional offset into the transform matrix when the transform block size is not square,
 wherein transforming the array of residual samples further comprises selecting the submatrix using the offset and the additional offset. 
 
     
     
       3. The method of  claim 1 , wherein transforming the array of residual samples further comprises decomposing the array of residual samples and the submatrix using even-odd decomposition. 
     
     
       4. The method of  claim 1 , wherein transforming the array of residual samples further comprises performing matrix multiplications between the submatrix and the array of residual samples using a matrix multiplication engine. 
     
     
       5. The method of  claim 4 , further comprising applying a directional transform to another array of residual samples to generate another array of transform coefficient values,
 wherein matrix multiplications between at least one directional transform matrix and the other array of residual samples are performed using the matrix multiplication engine. 
 
     
     
       6. A method comprising:
 computing an offset into a transform matrix based on a transform block size, wherein a size of the transform matrix is larger than the transform block size; 
 selecting a submatrix of the transform block size from the transform matrix using the offset; and 
 transforming an array of transform coefficient values based on the submatrix to reconstruct an array of residual samples. 
 
     
     
       7. The method of  claim 6 , further comprising:
 deriving an inverse quantization step size for the array of transform coefficient values; and 
 dequantizing each coefficient value in the array of transform coefficient values using the inverse quantization step size. 
 
     
     
       8. The method of  claim 6 , wherein the offset is a first offset, the method further comprising:
 computing an additional offset into the transform matrix when the transform block size is not square; and 
 transforming the array of transform coefficient values further comprises using the first offset and the additional offset to select the submatrix. 
 
     
     
       9. The method of  claim 6 , wherein the transform matrix includes 7-bit signed fixed point values. 
     
     
       10. The method of  claim 6 , wherein transforming the array of transform coefficient values further comprises decomposing the array of transform coefficient values and the submatrix using even-odd decomposition. 
     
     
       11. The method of  claim 6 , wherein transforming the array of transform coefficient values further comprises performing matrix multiplications between the submatrix and the array of transform coefficient values using a matrix multiplication engine. 
     
     
       12. The method of  claim 11 , further comprising applying an inverse directional transform to the array of transform coefficient values to reconstruct another array of residual samples,
 wherein matrix multiplications between at least one inverse directional transform matrix and the array of transform coefficient values are performed using the matrix multiplication engine. 
 
     
     
       13. A digital system comprising:
 an offset calculator configured to compute an offset into a transform matrix based on a transform block size, wherein a size of the transform matrix is larger than the transform block size, and wherein the transform matrix is a forward transform matrix or an inverse transform matrix; and 
 a transform engine configured to transform a residual block to generate a transform coefficient block, 
 wherein the offset is used to select elements of rows and columns of the submatrix of the transform block size from the transform matrix. 
 
     
     
       14. The digital system of  claim 13 , wherein the transform engine is further configured to decompose the array of residual samples and the submatrix using even-odd decomposition. 
     
     
       15. The digital system of  claim 13 , wherein the transform engine comprises a matrix multiplication engine and the transform engine is further configured to perform matrix multiplications between the submatrix and the array of residual samples using the matrix multiplication engine. 
     
     
       16. The digital system of  claim 13 ,
 wherein the transform engine comprises a matrix multiplication engine, and 
 wherein the transform engine is further configured to perform matrix multiplications for application of a directional transform to the array of residual samples using the matrix multiplication engine. 
 
     
     
       17. The method of  claim 6 , wherein computing the offset comprises computing the offset into the transform matrix in only one dimension. 
     
     
       18. The method of  claim 6 , wherein the transform matrix includes 8-bit signed fixed point values. 
     
     
       19. The method of  claim 6 , further comprising:
 generating a reconstructed video sequence based on the array of residual samples; and 
 presenting the reconstructed video sequence via a light emitting diode display. 
 
     
     
       20. The method of  claim 6 , further comprising:
 generating a reconstructed video sequence based on the array of residual samples; and 
 presenting the reconstructed video sequence via a liquid crystal display.